Do placebos work and if so is there in routine practice ethical?
The answer to the first question is still not clear. It is clear that in many placebo controlled trials patients given placebo improve. Whether this is just regression to the mean or a true effect of the placebo is unclear.
I see enough patients in whom treatments haven't worked to have much faith in the placebo effect for most illnesses. What I do wonder about is chronic illnesses with poorly defined biological basis like chronic pain syndromes, fibromyalgia, chronic fatigue syndrome, irritable bowel, etc. These patients clearly experience discomfort, but without a real biological basis there isn't much to go on for treatment. In these cases I suspect a placebo might help.
What I can't see is actually incorporating them into practice. How could I make a patient believe I was giving them a real treatment and then give them a placebo? What would I write on the prescription? I think if I lied to them (which is what I'd be doing) and they found out, they wouldn't trust me anymore, and I'm pretty sure I couldn't lie to them in good faith in the first place.